util.gen
Class SpearmanCorrelation

java.lang.Object
  util.gen.SpearmanCorrelation

public class SpearmanCorrelation

extends java.lang.Object

Calculates a Spearman correlation coefficient (rho), in the case of ties, averages the ranks and returns a Pearson correlation coefficient on the ranks.

Author:

Nix

Constructor Summary

SpearmanCorrelation()

Method Summary

double	corrCoeff(float[] a, float[] b) This is the final step in calculating a Spearman Correlation coeff.
RankedFloatArray	getARFA()
RankedFloatArray	getBRFA()
static void	main(java.lang.String[] args)
RankedFloatArray	rank(float[] f)
RankedFloatArray	rankSamples(RankSample[] rs) Ranks a sorted array of RankSample based on value.
void	setARFA(RankedFloatArray arfa)
void	setBRFA(RankedFloatArray brfa)
double	spearmanCorrelationCoefficient(float[] a, float[] b) Returns a Spearman Rank Correlation Coefficient (rho) between the two arrays.
double	spearmanCorrelationCoefficient(RankedFloatArray aRFA, RankedFloatArray bRFA) Returns a Spearman Rank Correlation Coefficient (rho) between the two arrays.

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

SpearmanCorrelation

public SpearmanCorrelation()

Method Detail

spearmanCorrelationCoefficient

public double spearmanCorrelationCoefficient(float[] a,
                                             float[] b)

Returns a Spearman Rank Correlation Coefficient (rho) between the two arrays. If ties are found, their ranks are averaged and a Pearson correlation is calculated on the ranks.

spearmanCorrelationCoefficient

public double spearmanCorrelationCoefficient(RankedFloatArray aRFA,
                                             RankedFloatArray bRFA)

Returns a Spearman Rank Correlation Coefficient (rho) between the two arrays. If ties are found, their ranks are averaged and a Pearson correlation is calculated on the ranks.

corrCoeff

public double corrCoeff(float[] a,
                        float[] b)

This is the final step in calculating a Spearman Correlation coeff. Don't use it unless you know what you are doing

Parameters:

a - - a values converted to ranked and sorted by original index

b - - ditto Call correlate() if you don't have the ranks.

rank

public RankedFloatArray rank(float[] f)

rankSamples

public RankedFloatArray rankSamples(RankSample[] rs)

Ranks a sorted array of RankSample based on value. If no ties are found then this is simply their array index number+1. (ie 1,2,3,4...) If ties are encountered, ties are assigned the average of their index positions+1. (ie if index+1's: 2,3,4 have the same absolute difference, all are assigned a rank of 3).

main

public static void main(java.lang.String[] args)

getARFA

public RankedFloatArray getARFA()

setARFA

public void setARFA(RankedFloatArray arfa)

getBRFA

public RankedFloatArray getBRFA()

setBRFA

public void setBRFA(RankedFloatArray brfa)